The primary driver of console sales is how many great games there are to play on the system. So, given the long shelf lives of Wii games compared to the flash-in-the-pan popularity of PlayStation 3 games, it's no surprise that the Wii is dominating the PS3 in sales.
But there are also aspects of the system itself that users pay attention to -- like, does it work? The most recent issue of Game Informer has a few charts that show how reliable each of the three current consoles are. They surveyed nearly 5000 people (I assume mostly through their website), and though this may introduce error into the absolute estimates (e.g., is the true percent of faulty PS3s 10% or 15%?), it shouldn't mess up the relative picture -- that is, which system screws up more than which other? Here are their results:
1. % whose consoles have failed: 54.2 for Xbox 360, 10.6 for PS3, 6.8 for Wii
2. % whose consoles failed again after first fix-up: 41.2 for Xbox 360, 14.7 for PS3, 11 for Wii
3. % whose friends have had these consoles fail: 69.9 for Xbox 360, 12.4 for PS3, 6 for Wii
4. % who said customer service was "very helpful": 37.7 for Microsoft, 51.1 for Sony, 56.1 for Nintendo
The writer offers a lame potential excuse that Wii owners may play them a lot less, but they don't provide statistics to support this. They say that about 40% of Wii owners play it for less than an hour a day, and that about 40% of Xbox 360 and PS3 owners play them for 3 - 5 hours a day. But this isn't the full distribution, and it doesn't say what the average owner does -- for all we know, 40% of Wii owners play it for less than an hour, while 60% play it for over 5 hours, and 40% of 360 and PS3 owners play them for 3 - 5 hours, while 60% play it for less than an hour. If stereotypes are true, most of the people for whom video games are a second full-time job are playing the 360 and PS3 -- but that doesn't mean the average owner of those consoles is so obsessed. (This mixes up "most X are Y" with "most Y are X".)
Still, even if it were true that Wii owners play it much less often, that can't account for the pattern because it's basically a 360 vs. not-360 distinction. Maybe less time playing accounts for the small difference between the Wii and PS3 failure rates, but not for the main pattern. Plus, the idea that playing a system more leads to greater breakdowns sounds like bullshit -- it's mostly due to how well the system is designed. The Atari, top-loader Nintendo, Super Nintendo, Sega Genesis, TurboGrafx-16 -- they all work, even though they've been put under stress for decades. The non-trivial fraction of Xbox 360s that have failed are less than 4 years old (at the oldest).
So, in addition to having more games with long-lasting appeal, the Wii hardware is least likely to break, and if something does go wrong, their customer service is the most helpful. Clearly the hardware and customer service aspects are not primary, since that would predict the PS3 vastly outselling the Xbox 360 and being neck-and-neck with the Wii. Still, these matters of secondary importance shouldn't be forgotten when we try to account for why the Wii is selling so well. The answer is pretty boring to normal people -- it's better in all key regards -- but is upsetting to most of the 20 and 30-something males who spend their lives shackled to their PS3.